Network Element Health Status Detection Method and Device

1. A network element health status detection method, comprising: determining first sampled data of at least one key performance indicator (KPI) of a target network element in a first time window; determining third sampled data of the at least one KPI in a third time window, wherein the third time window is greater than the first time window, and wherein a third end time of the third time window is less than or equal to a first end time of the first time window; obtaining a fluctuation score of the at least one KPI according to the first sampled data of the at least one KPI in the first time window and a steady state value of the at least one KPI; obtaining a trend score of the at least one KPI according to the third sampled data of the at least one KPI in the third time window; and determining, a health status of the target network element based on the fluctuation score and the trend score of the at least one KPI.

2. The method according to claim 1 , wherein the fluctuation score represents a degree of deviation of the at least one KPI from a steady state represented by the steady state value.

3. The method according to claim 1 , wherein obtaining the fluctuation score comprises: calculating a distance between the first sampled data of the at least one KPI in the first time window and the steady state value of the at least one KPI; and obtaining the fluctuation score according to the distance between the first sampled data of the at least one KPI in the first time window and the steady state value of the at least one KPI.

4. The method according to claim 3 , wherein calculating the distance comprises calculating the distance between the first sampled data of the at least one KPI in the first time window and the steady state value of the at least one KPI using one of a standard deviation method, an average deviation method, or a variation coefficient method.

5. The method according to claim 1 , further comprising: determining second sampled data of the at least one KPI in a second time window, wherein the second time window is greater than the first time window, and wherein a second end time of the second time window is less than or equal to the first end time of the first time window; and calculating the steady state value of the at least one KPI based on the second sampled data of the at least one KPI in the second time window.

6. The method according to claim 5 , wherein calculating the steady state value comprises calculating the steady state value of the at least one KPI based on the second sampled data of the at least one KPI in the second time window after at least one of an abnormal point or a noise point is removed.

7. The method according to claim 1 , wherein determining the health status of the target network element comprises: obtaining a reliability score of the at least one KPI through weighted calculation according to the fluctuation score and the trend score of the at least one KPI; and determining the health status of the target network element based on the reliability score of the at least one KPI.

8. The method according to claim 7 , further comprising: determining fourth sampled data of the at least one KPI of the target network element in a detection moment, wherein the detection moment is the first end time of the first time window; determining a score of a distance between the at least one KPI and a network element hardware property threshold based on the fourth sampled data and a preset network element hardware property threshold of the at least one KPI; and determining the health status of the target network element based on the fluctuation score, the trend score of the at least one KPI, and the score of the distance between the at least one KPI and the network element hardware property threshold.

9. The method according to claim 8 , further comprising: decomposing the third sampled data of the at least one KPI in the third time window to determine a trend component of the at least one KPI; and obtaining the trend score of the at least one KPI according to the trend component of the at least one KPI.

10. The method according to claim 7 , wherein the at least one KPI comprises a plurality of KPIs, and wherein determining the health status of the target network element comprises: performing weighted calculation on reliability scores of each of the plurality of KPIs to determine a second reliability score of the target network element; and determining the health status of the target network element based on the second reliability score of the target network element.

11. The method according to claim 1 , wherein the health status of the target network element comprises a fault state, a sub-healthy state, or a normal state.

12. The method according to claim 1 , wherein the at least one KPI comprises a pre-correction bit error rate or a post-correction bit error rate.

13. A network element health status detection apparatus, comprising: a processor; and a non-transitory computer readable medium containing computer-executable instructions executable by the processor such that when executed, enable the network element health status detection apparatus to: determine first sampled data of at least one key performance indicator (KPI) of a target network element in a first time window; determine third sampled data of the at least one KPI in a third time window, wherein the third time window is greater than the first time window, and wherein a third end time of the third time window is less than or equal to a first end time of the first time window; obtain a fluctuation score of the at least one KPI according to the first sampled data of the at least one KPI in the first time window and a steady state value of the at least one KPI; obtain a trend score of the at least one KPI according to the third sampled data of the at least one KPI in the third time window; and determine a health status of the target network element based on the fluctuation score and the trend score of the at least one KPI.

14. The apparatus according to claim 13 , wherein the fluctuation score represents a degree of deviation of the at least one KPI from a steady state represented by the steady state value.

15. The apparatus according to claim 13 , wherein the apparatus obtains the fluctuation score by: calculating a distance between the first sampled data of the at least one KPI in the first time window and the steady state value of the at least one KPI; and obtaining the fluctuation score according to the distance between the first sampled data of the at least one KPI in the first time window and the steady state value of the at least one KPI.

16. The apparatus according to claim 15 , wherein calculating the distance comprises the apparatus calculating the distance between the first sampled data of the at least one KPI in the first time window and the steady state value of the at least one KPI using one of a standard deviation method, an average deviation method, or a variation coefficient method.

17. The apparatus according to claim 13 , wherein the processor is further configured to execute the computer-executable instructions to enable the apparatus to: determine second sampled data of the at least one KPI in a second time window, wherein the second time window is greater than the first time window, and wherein a second end time of the second time window is less than or equal to the first end time of the first time window; and calculate the steady state value of the at least one KPI based on the second sampled data of the at least one KPI in the second time window.

18. The apparatus according to claim 17 , wherein the apparatus calculates the steady state value by calculating the steady state value of the at least one KPI based on the second sampled data of the at least one KPI in the second time window after at least one of an abnormal point or a noise point is removed.

19. The apparatus according to claim 13 , wherein the apparatus determines the health status of the target network element by: obtaining a reliability score of the at least one KPI through weighted calculation according to the fluctuation score and the trend score of the at least one KPI; and determining the health status of the target network element based on the reliability score of the at least one KPI.

20. The apparatus according to claim 19 , wherein the processor is further configured to execute the computer-executable instructions to enable the apparatus to: determine fourth sampled data of the at least one KPI of the target network element in a detection moment, wherein the detection moment is the first end time of the first time window; determine a score of a distance between the at least one KPI and a network element hardware property threshold based on the fourth sampled data of the at least one KPI in the detection moment and a preset network element hardware property threshold of the at least one KPI; and determine the health status of the target network element based on the fluctuation score, the trend score of the at least one KPI, and the score of the distance between the at least one KPI and the network element hardware property threshold.

21. The apparatus according to claim 20 , wherein the processor is further configured to execute the computer-executable instructions to enable the apparatus to: decompose the third sampled data of the at least one KPI in the third time window to determine a trend component of the at least one KPI; and obtain the trend score of the at least one KPI according to the trend component of the at least one KPI.

22. The apparatus according to claim 19 , wherein the at least one KPI comprises a plurality of KPIs, and wherein the apparatus determines the health status of the target network element by: performing weighted calculation on reliability scores of each of the plurality of KPIs to determine a second reliability score of the target network element; and determining the health status of the target network element based on the second reliability score of the target network element.

23. The apparatus according to claim 13 , wherein the health status of the target network element comprises one of a fault state, a sub-healthy state, or a normal state.